Dirk S. Paul

Human genetic studies have identified key regulators of fetal haemoglobin (HbF) expression, including BCL11A, resulting in therapeutic advances ^1–8 . Yet the mechanisms by which HbF expression is activated remain incompletely understood ⁹ . Here we conduct a large multi-ancestry genome-wide association study of HbF levels in 28,279 individuals that identifies 91 conditionally independent associations across 12 genomic regions. In one previously uncharacterized associated region, the high-HbF-linked causal variant rs1010474-C reduces BACH2 expression and elevates HbF levels. Direct perturbation or inhibition of BACH2 likewise increases HbF expression. Mechanistically, BACH2 restrains activation of the HbF-encoding γ-globin genes, while loss of BACH2 enhances NRF2 chromatin occupancy and promotes the formation of activation foci at the γ-globin genes. Although BACH2 and NRF2 binding motifs in the γ-globin promoters overlap, they can be selectively edited to activate or repress γ-globin, respectively, and do so independently of BCL11A. These findings illustrate how human genetic variation continues to advance our understanding of therapeutically relevant regulatory mechanisms underlying HbF expression.

Many non-coding variants influence complex traits and diseases through gene regulation, yet the mechanisms linking these variants to downstream biology remain poorly understood. Here, we present eQTLGen Phase 2 , a comprehensive genome-wide analysis of gene expression quantitative trait loci (eQTLs) in 43,301 blood samples from 52 datasets. Beyond local cis -effects, this sample size enabled the first systematic mapping of trans -eQTLs at scale. We identify cis -eQTLs for nearly all expressed genes (94.7%) and trans -eQTLs for over half (56.2%). Second, by colocalizing cis -eQTLs with trans -eQTLs, we infer a directed gene regulatory network comprising 47,554 directed gene regulatory relationships. These networks reveal how genetic perturbations in upstream regulators produce dose-dependent downstream effects, supported by Perturb-seq and ChIP-seq data. Third, integrating this network with 87 genome-wide association studies allows us to systematically prioritize trait-relevant pathways and candidate genes. Variants exerting both cis – and trans -effects are markedly more likely to colocalize with trait associations than cis -only variants, delineating a subset of functionally active cis -eQTLs from a large group with limited downstream impact. This distinction provides a conceptual framework for identifying regulatory variants that truly mediate complex trait biology. Together, these results provide a publicly available resource of cis – and trans -eQTLs and an in vivo scaffold for human gene-regulatory networks, elucidating how propagation of cis -effects modulates complex disease.

The biological mechanisms through which most nonprotein-coding genetic variants affect disease risk are unknown. To investigate gene-regulatory mechanisms, we mapped blood gene expression and splicing quantitative trait loci (QTLs) through bulk RNA sequencing in 4,732 participants and integrated protein, metabolite and lipid data from the same individuals. We identified cis -QTLs for the expression of 17,233 genes and 29,514 splicing events (in 6,853 genes). Colocalization analyses revealed 3,430 proteomic and metabolomic traits with a shared association signal with either gene expression or splicing. We quantified the relative contribution of the genetic effects at loci with shared etiology, observing 222 molecular phenotypes significantly mediated by gene expression or splicing. We uncovered gene-regulatory mechanisms at disease loci with therapeutic implications, such as WARS1 in hypertension, IL7R in dermatitis and IFNAR2 in COVID-19. Our study provides an open-access resource on the shared genetic etiology across transcriptional phenotypes, molecular traits and health outcomes in humans ( https://IntervalRNA.org.uk ).

Genomics can provide insight into the etiology of type 2 diabetes and its comorbidities, but assigning functionality to non-coding variants remains challenging. Polygenic scores, which aggregate variant effects, can uncover mechanisms when paired with molecular data. Here, we test polygenic scores for type 2 diabetes and cardiometabolic comorbidities for associations with 2,922 circulating proteins in the UK Biobank. The genome-wide type 2 diabetes polygenic score associates with 617 proteins, of which 75% also associate with another cardiometabolic score. Partitioned type 2 diabetes scores, which capture distinct disease biology, associate with 342 proteins (20% unique). In this work, we identify key pathways (e.g., complement cascade), potential therapeutic targets (e.g., FAM3D in type 2 diabetes), and biomarkers of diabetic comorbidities (e.g., EFEMP1 and IGFBP2) through causal inference, pathway enrichment, and Cox regression of clinical trial outcomes. Our results are available via an interactive portal (https://public.cgr.astrazeneca.com/t2d-pgs/v1/).

The G haplotype is a group of co-inherited single nucleotide variants in the F5 gene that reduce venous thromboembolism (VTE) risk. Although 7% of the population is homozygous for the G haplotype (F5-G/G), the underlying mechanism of VTE protection is poorly understood. Using RNA sequencing data from 4651 blood donors in the INTERVAL study, we detected a rare excision event at the factor V (FV)-short splice sites in 5% of F5-G/Gs carriers as compared with 2.16% of homozygotes for the F5 reference sequence (F5-ref; P = .003). Highly elevated (∼10-fold) FV-short, a FV isoform that lacks most of the B-domain, has been linked with increased tissue factor inhibitor α (TFPIα) levels in rare hemorrhagic diathesis, including East Texas bleeding disorder. To ascertain whether the enhanced FV-short splicing seen in F5-G/G INTERVAL participants translated to increased plasma FV-short levels, we analyzed plasma samples from 7 F5-G/G and 13 F5-ref individuals in a recall-by-genotype study. A ∼2.2-fold higher amount of FV-short was found in a plasma pool from F5-G/G participants when compared with the pool of F5-refs (P = .029), but there was no difference in the total FV levels. Although no significant difference in TFPI levels were found, F5-G/Gs showed a ∼1.4-fold TFPI-dependent increase in lag time to thrombin generation than F5-refs (P = .0085). Finally, in an analysis of 117 699 UK Biobank participants, we discovered that, although being protective against VTE, the G haplotype also confers an increase in bleeding episodes (P = .011). Our study provides evidence that the effect of the common G haplotype is mediated by the FV-short/TFPI pathway.

Key Points: We conducted a clinical, genetic, and pathological analysis on 64 cases from 39 families with TRPC6-associated podocytopathy (TRPC6-AP). Analysis of 37,542 individuals excluded a major contribution of loss-of-function variants to TRPC6-AP, legitimating current drug discovery approaches. This study identifies key features of disease that can help intervention studies design and suggests similarities between TRPC6-AP and primary FSGS. Background: Understanding the genetic basis of human diseases has become integral to drug development and precision medicine. Recent advancements have enabled the identification of molecular pathways driving diseases, leading to targeted treatment strategies. The increasing investment in rare diseases by the biotech industry underscores the importance of genetic evidence in drug discovery and approval processes. Here we studied a monogenic Mendelian kidney disease, TRPC6-associated podocytopathy (TRPC6-AP), to present its natural history, genetic spectrum, and clinicopathological associations in a large cohort of patients with causal variants in TRPC6 to help define the specific features of disease and further facilitate drug development and clinical trials design. Methods: The study involved 64 individuals from 39 families with TRPC6 causal missense variants. Clinical data, including age of onset, laboratory results, response to treatment, kidney biopsy findings, and genetic information, were collected from multiple centers nationally and internationally. Exome or targeted sequencing was performed, and variant classification was based on strict criteria. Structural and functional analyses of TRPC6 variants were conducted to understand their effect on protein function. In-depth reanalysis of light and electron microscopy specimens for nine available kidney biopsies was conducted to identify pathological features and correlates of TRPC6-AP. Results: Large-scale sequencing data did not support causality for TRPC6 protein-truncating variants. We identified 21 unique TRPC6 missense variants, clustering in three distinct regions of the protein, and with different effects on TRPC6 3D protein structure. Kidney biopsy analysis revealed FSGS patterns of injury in most cases, along with distinctive podocyte features including diffuse foot process effacement and swollen cell bodies. Most patients presented in adolescence or early adulthood but with ample variation (average 22, SD ±14 years), with frequent progression to kidney failure but with variability in time between presentation and kidney failure. Conclusions: This study provides insights into the genetic spectrum, clinicopathological associations, and natural history of TRPC6-AP. Clinical Trial registry name and registration number: A Study to Test BI 764198 in People With a Type of Kidney Disease Called Focal Segmental Glomerulosclerosis, NCT05213624.

Liver X receptor-α (LXRα) regulates cellular cholesterol abundance and potently activates hepatic lipogenesis. Here we show that at least 1 in 450 people in the UK Biobank carry functionally impaired mutations in LXRα, which is associated with biochemical evidence of hepatic dysfunction. On a western diet, male and female mice homozygous for a dominant negative mutation in LXRα have elevated liver cholesterol, diffuse cholesterol crystal accumulation and develop severe hepatitis and fibrosis, despite reduced liver triglyceride and no steatosis. This phenotype does not occur on low-cholesterol diets and can be prevented by hepatocyte-specific overexpression of LXRα. LXRα knockout mice exhibit a milder phenotype with regional variation in cholesterol crystal deposition and inflammation inversely correlating with steatosis. In summary, LXRα is necessary for the maintenance of hepatocyte health, likely due to regulation of cellular cholesterol content. The inverse association between steatosis and both inflammation and cholesterol crystallization may represent a protective action of hepatic lipogenesis in the context of excess hepatic cholesterol.

Understanding the function of genetic variants associated with human traits and diseases remains a significant challenge. Here we combined analyses based on natural genetic variation and genetic engineering to dissect the function of 94 non-coding variants associated with haematological traits. We describe 22 genetic variants impacting haematological variation through gene expression. Further, through in-depth functional analysis, we illustrate how a rare, non-coding variant near the CUX1 transcription factor impacts megakaryopoiesis through modulation of the CUX1 transcriptional cascade. Collectively, our findings enhance the functional interpretation of genetic association studies and advance understanding of how non-coding variants contribute to blood and immune system variation.

Obesity is a major risk factor for many common diseases and has a substantial heritable component. To identify new genetic determinants, we performed exome-sequence analyses for adult body mass index (BMI) in up to 587,027 individuals. We identified rare loss-of-function variants in two genes (BSN and APBA1) with effects substantially larger than those of well-established obesity genes such as MC4R. In contrast to most other obesity-related genes, rare variants in BSN and APBA1 were not associated with normal variation in childhood adiposity. Furthermore, BSN protein-truncating variants (PTVs) magnified the influence of common genetic variants associated with BMI, with a common variant polygenic score exhibiting an effect twice as large in BSN PTV carriers than in noncarriers. Finally, we explored the plasma proteomic signatures of BSN PTV carriers as well as the functional consequences of BSN deletion in human induced pluripotent stem cell-derived hypothalamic neurons. Collectively, our findings implicate degenerative processes in synaptic function in the etiology of adult-onset obesity.

Introduction Type 2 diabetes (T2D) is a heterogeneous disorder for which disease-causing pathways are incompletely understood. Here, we mapped genetic risk for T2D and its comorbidities to proteins, mechanistic pathways and clinical outcomes using proteogenomic data from a population-scale biobank and two randomized controlled trials. Methods We tested polygenic scores (PGS) for T2D and its cardiometabolic comorbidities, plus five partitioned T2D PGS (beta cell, lipodystrophy, liver lipid, obesity, and liver lipid), for association with 2,922 circulating proteins in 54,306 multi-ancestry participants (of which 42,452 were unrelated and without prevalent cardiometabolic disease) from the UK Biobank (UKB). Then, we tested the PGS-associated proteins for association with incident cardiometabolic complications in two cardiovascular outcome trials among T2D patients with proteogenomic data: EXSCEL (N=2,823) and DECLARE-TIMI 58 (N=915). We assessed causality using two-sample Mendelian randomization and mediation. Results We identified 839 unique proteins significantly associated with any T2D PGS and 1,005 proteins that were associated with at least one cardiometabolic PGS. Some PGS-associated proteins such as TFF3, EFEMP1, and MMP12 were in turn associated with renal and cardiovascular trial outcomes. PGS association patterns revealed shared pathways, e.g., complement cascade, cholesterol metabolism, IGF signaling. The proteins underlying these pathways, such as LPA, C1S, and IGFBP2, were consistently associated with clinical trial outcomes or identified via causal inference. Conclusions This proteogenomic study revealed proteins and mechanistic pathways underlying T2D and related comorbidities, advancing our understanding of T2D pathobiology and identifying putative biomarkers. All our results are available in an online data portal (https://public.cgr.astrazeneca.com/t2d-pgs/v1/).